Understood. I’m just treating today as the datum point, it is what it is.
1 Like
I’m very sorry for my overly harsh words. I really like this series.
2 Likes
This is probably why my view of 2024 is a very positive one - I’m on gamepass regardless as it’s how I fly on 2020 and, along with my son, play other games. So 2024 hasn’t cost me a penny and I’m happily flying the new A330 on free flight with no real complaints (aside from static aircraft which I complain about on here daily 
)
2 Likes
Well. . . Thank you!
I’ve worked in both hardware and software development and testing for, (
), decades, and though I may not have “seen it all”, I think I’ve seen enough to know which end of the soldering-iron gets hot and which end of the keyboard faces up.
Because of this, I have a pretty good idea how difficult a software project of any complexity can be - and that the difficulty rises in roughly an exponential function of complexity.
I also know that in every software project more complex than a “Hello World!” program, there are “gotchas” that seem to come out of the woodwork that NOBODY saw coming[1]. (And even a simple “Hello World!” program in pure amd64 assembler, designed to run in protected mode, is non-trivial. Go ahead, ask me how I know.)
I suspect, (but cannot prove), that Jorg & Co. had a different vision from the rushed monstrosity that was released - perhaps a more nuanced approach where they got a refactored baseline working reasonably well, then released individual features as updates - but were overruled by The Powers That Be into throwing stuff together and hoping it sticks.
Am I disappointed? Yes.
Do I blame Jorg & Co., the people stuck in the middle of this mess? No.
My take on this, (based on my own decades of experience), is “Walk a mile in my shoes”[2]. We really don’t know what’s going on behind closed doors so we should be gracious - at least with respect to the developers.
What say ye?
==================== Footnotes ====================
-
I decided to write a “simple” modification to an existing program for the GoPiGo-3 robot. The initial program was called “remote camera robot” and used a mouse to drag a dot around the screen that controls the movement of the robot, and has a FPV camera view so you can see where you’re going.
This program was difficult to use and needed serious tweaking to make it even minimally controllable, so I decided to modify it to use a joystick instead of the kludgy mouse-and-dot. How difficult could that possibly be? Right?
Answer: Incredibly difficult, even though something like 90% of the code was reusable as-is.
I ended up with a crash-course in browser-based client-server models in both Python and JavaScript. I also discovered that some genius had decided that any use of any game controller in any browser, anywhere, should be required to have a real, live, domain and a real, live, traceable to a root authority, site certificate for that domain, even if the robot was its own access point broadcasting a private IP address. (10.10.10.10). Which resulted in another crash course in NGINX and secure servers.
I don’t blame the developers at all. . . . -
One of the comments on YouTube was this:
4 Likes
But it was released, and we are where we are. I don’t see what benefit there is in continuing to bring this up. I’d wager that Asobo knew full well it wasn’t ready, but the MS coffers needed satiating.
It’s been oft stated that Asobo aren’t sufficiently qualified for a project of this magnitude — and that maybe someone like Rockstar (just as an example — I’m not clued up on the world of games) would have done a better programming/coding job — however MSFS isn’t “just another game.” I believe we have the right people for the task. You can be as competent as you like in software development but without a passion for and knowledge of the subject of the product, that’s probably immaterial.
No one’s forcing anyone to buy FS2024. No one’s scamming or has scammed anyone. And by definition, we do not have a totally broken simulator. I get that people are having problems but, conversely, others (like me) are running it with only the already-identified assortment of (presumably) fixable bugs.
6 Likes
Personal comments:
You obviously weren’t there for the launch of Red Dead Redemption 2 
I don’t really think that’s fair to Asobo. If they successfully fix bugs, like all the fixes in SU1, then logically they are ‘sufficiently qualified’. It’s just a matter of time constraints for the initial launch.
If the Sim had launched a couple of months later but with half the SU1 fixes, it would have had a much better reception. If they keep pushing SU1 back then that implies that the Sim updates aren’t set in stone, but if they didn’t push the initial release back that implies that the inital release was. Bugs aren’t magically fixed so either you push the release back or release with bugs.
I don’t think there are any programmers in existance that don’t slip up and write buggy software.
5 Likes
There have been many fixes not documented in the release notes and as I recall all fixes will be documented when SU1 final is released.
1 Like
You make it sound like there are only 10, 100, or even 1,000 issues with Asobo’s code. Not 18,000 or 50,000. To write that many bugs into one program you’d have to be doing it intentionally.
And you are correct, most programmers do make mistakes. The difference is that they are quick to fix them and are very vocal about it.
If I told my boss I had a program with 50,000 bugs ready for release, I’d be getting a pink slip.
2 Likes
The big difference here, (IMHO), is that it was the bosses who told them to ship it.
6 Likes
I’d guess the bosses need the pink note then…
3 Likes
Let’s hope that doesn’t happen or we will all be SOL…lol.
4 Likes
In the movie The Good, The Bad, and The Ugly, (AFAIK), The character portrayed by Clint Eastwood said:
The bosses had the big guns and the dev’s dug.
4 Likes
Personal Comments and Observations
I wear several hats in addition to being Volunteer Staff here. One of them is working for a Dev Partner whose name is on the Marquee board when FS’24 boots up. So I have a fairly unique view into how the sausage is made so to speak having to live here and see / experience Customer Sentiment and then actually building in the sim and interacting with the First Parties (MS Asobo).
Asobo is more than qualified. The living proof is FS2020. What most end users won’t ever appreciate until they have to live in it is that nearly everything that encompasses the sim experience is centered around the SDK. Without it, the sim doesn’t exist. But like many codexes, it didn’t spring into existence by itself. A deliberate, detailed process to create a Framework by which flight simulation could be translated into working code is no easy feat. And it’s a Living Document - it is never done. So while a lot of it is stable, a lot of it is still being matured. Asobo built on it and improved/innovated it.
The other piece is the complexity - the sim by itself can’t be run solely anymore on a local PC. Those days are long gone. The sheer Petabytes of data to render the Digital Twin World puts paid to that. So many players who are used to standalone PC apps are suddenly thrust into a world more familiar to online players of MMORPGs like WoW. So there’s this cultural shock that hits the user base as well.
Pointing the finger at one First Party is to deny the factors above (which are by no means all encompassing - I’ve barely scratched the surface). MS is qualified to provide the Infrastructure, Publication and Marketing, Asobo is qualified having rewritten the SDK from almost the ground up and delivered FS2020. They are supported by all the Dev Partners, past and present in both Sims.
'Nuff said as Stan Lee would say.
9 Likes
I reckon quite the opposite. I surmise SU1 is 100% done apart from Marketplace which is 99.999999% done. And Jorg is standing on saying “SHOW ME IT’S PERFECT!!!”
(Quoting the whole passage for context)
Because Marketplace has to be perfect first time doesn’t it.
1 Like
I’m just quoting what I’ve frequently read in these forums. It doesn’t reflect my own opinion. In fact, I’d say the version of FS2020 we’re all now playing is testament to Asobo’s ability. A flight sim isn’t just about programming — it’s about physics, aerodynamics, avionics, etc. You can have the best game developer in the world but if they don’t understand the science behind powered flight…
In my opinion, a game dealing with flight simulation is best created by those with an active interest in aviation.
1 Like
Preach it brother, preach it!
I think Theodore Roosevelt said it perfectly:
We should all print this and put it next to our sim rigs.
4 Likes
With all the money involved yes, the Marketplace has to be perfect. Honestly, I think Jorg is a little nervous releasing it.
Regarding SU1, today is Dev Blog day. I am hoping they will indicate it’s time to move on to SU2.
3 Likes
Last time I was coding Direct3D apps, we weren’t calling software a “beta” until it had pretty much all of its specified features, and most of the glaring issues had also been resolved. This was a bug hunting phase. Yeah, there might be lots of bugs and some would escape of course.
“Alpha” was used more for software that was in active development and had major features that were still missing. There might be some major issues with core features as well. Generally you didn’t give the customer alpha builds because there could be major stability issues as well. If you did give out builds like this, it was explicitly labeled as such so that the customer knew to expect some of this.
And so when I look at Flight Simulator 2024, which of those two best describes this? If you fly mainly in Free Flight with a well established aircraft, it might look like a beta. Depending on how and what you fly, it works for the most part, but you’ll find some bugs. However, this is mainly 2020 with some incremental upgrades to the color rendition, lighting, and flight planning. You’d expect it to at least be at that level.
If you look at 2024’s major new features like career mode, however, the software looks a lot more like an alpha. Large sections of the Career Mode are missing because they’re still in active development. There’s a lot shown on that opening loading screen video that you can’t do in the simulator yet. Sections of the career mode are indeed broken with an abundance of mission ruining bugs. And there are so many bugs you don’t hunt for them, you literally trip over them.
The streaming model also has major deficiencies that I’d say invalidate it from getting a beta label yet. Even with 24GB of VRAM and 64GB of RAM and a 200GB cache on the SSD, I’m still losing gauges midflight and runway textures that I’ve already downloaded more than once.
This doesn’t impact every user, and it shouldn’t be a system limitation in this case either, but it instead appears to because there are use cases for the software which simply haven’t been addressed yet. This is a core feature with a regular use case that isn’t working, and not addressing common use cases is also alpha or pre-alpha level software.. I’m really reluctant to say it’s because they haven’t tested all of these use cases because A. Many of them are extremely basic and B. they actually have a fairly large Q&A team according to their own accounts. It makes me think it’s far more likely that they know all about this already, and they willfully shipped in this state without notifying the customer.
4 Likes
Yes, for sure. Who seriously doubted that?
Did anyone really expect the “Next Generation of Flightsimulation” not to be better than the last? By “better” each of us certainly expected something different, but definitely not this: one step forward, three steps back.
I’m still on 2020, and the ongoing mess, and the obvious diversion of resources to fix the dumpster fire of 2024, I’ve basically stopped flying.
Brazil was supposed to be released 4 months ago. I’ll try the sim again this weekend, but the joy is gone. If they were to shut down tomorrow it would be a relief - no more gaslighting, misdirection, and bs.
2 Likes